the antitrust source
䡵 www.antitrustsource.com
䡵 October 2010
1
Including Exclusion in the 2010 Horizontal Merger
Guidelines
D . Bru ce H of fma n a n d Dan iel Francis
O
On August 19, 2010, the U.S. Department of Justice and the Federal Trade Commission issued
revised Horizontal Merger Guidelines.1 The 2010 Guidelines replace a 1992 version that was par-
tially amended in 1997.2 The new version differs markedly from its predecessor. Among other
things, it signals a reduced focus on market definition and de-emphasizes bright-line tests. It also
treats in detail a number of topics that were not extensively discussed in the 1992 Guidelines, and
places heightened emphasis on particular microeconomic tools.
One significant change has not received much attention to date: the introduction into the
Guidelines of an exclusionary analysis derived from Section 2 of the Sherman Act, which prohibits
monopolization via exclusionary conduct. This change appears at several places in the new
Guidelines. In particular:
● the list of statutes enforced by the Agencies through merger control now includes Section 2
of the Sherman Act 3;
● the “Overview” section now states that “[e]nhanced market power may also make it more
likely that the merged entity can profitably and effectively engage in exclusionary conduct”;
● a new section on sources of evidence now asserts that “rival firms may provide relevant
facts, and even their overall views may be instructive, especially in cases where the
Agencies are concerned that the merged entity may engage in exclusionary conduct” 4; and
● the discussion of unilateral effects focuses on price, capacity, and innovation, but cautions
that “[t]hese effects do not exhaust the types of possible unilateral effects; for example,
exclusionary unilateral effects also can arise.” 5
Introducing into merger review a test for potential future exclusionary conduct might seem like
an efficient “two-for-one” in cases where exclusionary concerns emerge during the merger control process. On closer examination, however, the new test likely cannot work. Moreover, we neither need such a test, nor—in light of its significant costs—would we want it.
Section 2: Treading New Ground in Merger Review
䡵
Section 2 of the Sherman Act, which prohibits monopolization and attempted monopolization, lies
D. Bruce Hoffman is a
at the heart of the antitrust laws. Properly interpreted and applied, Section 2 provides a valuable
partner with Hunton &
Williams and served as
1
Deputy and Associate
See U.S. Dep’t of Justice & Fed. Trade Comm’n, Horizontal Merger Guidelines (2010) [hereinafter 2010 Guidelines], available at http://
www.justice.gov/atr/public/guidelines/hmg-2010.pdf.
Director of the FTC’s
2
U.S. Dep’t of Justice & Fed. Trade Comm’n, Horizontal Merger Guidelines (1992, rev. 1997) [hereinafter 1992 Guidelines], available at
3
2010 Guidelines, supra note 1, § 1 (“The relevant statutory provisions include Section 7 of the Clayton Act, 15 U.S.C. § 18, Sections 1 and
Bureau of Competition
from 2001–2004.
http://www.justice.gov/atr/public/guidelines/hmg.htm. The 1997 amendment applied only to the 1992 Guidelines’ efficiencies section.
Daniel Francis is an
2 of the Sherman Act, 15 U.S.C. §§ 1, 2, and Section 5 of the Federal Trade Commission Act, 15 U.S.C. § 45.”).
associate with Hunton
4
Id. § 2.2.3.
& Williams.
5
Id. § 6.
the antitrust source
䡵 www.antitrustsource.com
䡵 October 2010
2
bulwark against certain kinds of anticompetitive exclusionary conduct and—although the boundaries of liability under monopolization law can be difficult to describe—it deserves vigorous,
thoughtful enforcement. Nevertheless, in our view, the merger review process is not the proper
forum for that enforcement.
As a historical matter, Section 2 has never been included in the list of statutory provisions that
To that end, the 2010
the Agencies have purported to enforce through merger control.6 Moreover, to the best of our
knowledge the Agencies have never challenged a horizontal merger that was otherwise unob-
Guidelines demand an
jectionable on the ground that the firm, post-merger, might be more likely or better able to engage
in exclusionary conduct in violation of Section 2.
It is true that in reviewing a relatively small subset of mergers, the Agencies have considered
inquiry into whether the
some very specific forms of exclusionary conduct. Thus, in vertical or partially vertical mergers—
merger would increase
i.e., where the merging firms are not just competitors but also have operations at different levels
of the supply chain—the Agencies have considered whether particular mergers might cause the
the ability and/or incen-
merged firms to deny inputs or outputs to competitors post-merger, and have sometimes shaped
merger remedies to address this concern.7 But this relatively uncommon situation is a vertical
tive of the merged firm
issue, not a horizontal one (and, indeed, was addressed in the Non-Horizontal Merger Guidelines 8 ). The new text seems to indicate that the Agencies have something quite different—and
to engage in exclusion-
more expansive—in mind: specifically, exclusionary effects that are not directly or immediately
linked to increased prices, reduced output, or constrained innovation. To that end, the 2010
ary conduct at some
Guidelines demand an inquiry into whether the merger would increase the ability and/or incentive
of the merged firm to engage in exclusionary conduct at some unspecified future time. We think
unspecified future time.
this approach raises intractable problems.
We think this approach
the difficulty of identifying situations in which existing Section 7 law could not reach competitive
An Intractable Task. Two aspects of the 2010 Guidelines’ new inquiry make it unmanageable:
concerns that would be caught by a Section 2 style “exclusion” analysis; and, the difficulty of
raises intractable
defining and detecting “exclusionary conduct” in this context.
problems.
that: (a) the merger will likely enable the merged firm (either by itself or in coordination with rivals)
First, under Section 7, the Agencies can already block mergers where the evidence shows both
to harm consumers by raising prices, reducing output, or forestalling innovation; and (b) those
potential harms outweigh any efficiencies from the merger.9 Thus, in order for an “exclusionary”
concern to arise that could not be adequately addressed under the 1992 Guidelines, one of two
6
See 1992 Guidelines, supra note 2, § 0 (“These Guidelines outline the present enforcement policy of the Department of Justice and the
Federal Trade Commission . . . concerning horizontal acquisitions and mergers . . . subject to section 7 of the Clayton Act, to section 1 of
the Sherman Act, or to section 5 of the FTC Act.”); U.S. Department of Justice, Non-Horizontal Merger Guidelines § 1.0 ( 1984) [hereinafter
1984 Guidelines], available at http://www.justice.gov/atr/public/guidelines/2614.htm (“These Guidelines state in outline form the present
enforcement policy of the U.S. Department of Justice . . . concerning acquisitions and mergers . . . subject to section 7 of the Clayton Act
or section 1 of the Sherman Act.”).
7
See, e.g., Press Release, U.S. Dep’t of Justice, Justice Department Requires Northrop Grumman to Adopt Non-Discrimination Terms in Order
to Consummate its Acquisition of TRW, Inc. (Dec. 11, 2002), available at http://www.justice.gov/atr/public/press_releases/2002/200543.pdf;
Press Release, Fed. Trade Comm’n, FTC Approves AOL/Time Warner Merger with Conditions, (Dec. 14, 2000), available at http://www.
ftc.gov/opa/2000/12/aol.shtm; Press Release, Fed. Trade Comm’n, FTC Settlement with Cadence, Cooper & Chyan to Preserve Competition
in Automated Chip Design Software (May 8, 1997), available at http://www.ftc.gov/opa/1997/05/cadence.shtm.
8
1984 Guidelines, supra note 6, § 4.2 (noting the risks of competitive harm associated with mergers that require simultaneous entry into
upstream and downstream markets, and indicating that “the Department will consider the likelihood of predatory price or supply ‘squeezes’
by the integrated firms against their unintegrated rivals”).
9
See generally, e.g., U.S. Dep’t of Justice and Fed. Trade Comm’n, Commentary on the Horizontal Merger Guidelines 2 (2006), available at
http://www.justice.gov/atr/public/guidelines/215247.pdf.
the antitrust source
䡵 www.antitrustsource.com
䡵 October 2010
3
situations would have to exist. The merged firm would either: (a) lack the power to raise price,
reduce output, or reduce innovation; or (b) have this power, but enjoy efficiencies of such magnitude that any loss of competition would be offset by the firm’s lower cost structures and the
resulting consumer benefits. Put differently, the new exclusion test could come into play only when
the merger somehow resulted in a merged firm that did not obtain market power (or that harnessed efficiencies sufficient to offset that power), but nevertheless gained some ability or incentive to engage, at some point in the future, in some form of exclusionary conduct that would, at
some even more distant future date, allow it to acquire or maintain monopoly power.
Second, exclusionary conduct is notoriously difficult to define even after the fact (or indeed
while the conduct is actually in progress), when at least we know what the putative monopolist is
accused of doing.10 Many of the practices that cause the most difficulties for antitrust courts under
Section 2 offer short-run consumer benefits and help to drive down costs and yet, in certain circumstances, may drive rivals from the market and ultimately inflict net harm on consumers. Due
in large part to these mixed effects, antitrust scholarship has yet to reach a consensus on where
the line should be drawn between vigorous competition which benefits the consumer and anticompetitive exclusion which does not—and all this is quite apart from the difficulties of working
out whether that line has been crossed in any individual case. In light of the significant practical
and doctrinal difficulties in applying Section 2 to concrete facts, evaluating the effects and legality of hypothetical future conduct begins to look implausible.
Accordingly, the new test calls on the Agencies to challenge a class of mergers that would be
procompetitive, or at worst neutral, under traditional Section 7 analysis for fear that the merged
firm might engage in conduct that no one can readily identify and might or might not be harmful
to consumers. We submit that if this can be done at all—and it probably cannot—it certainly cannot be done without a prohibitive risk of errors. There is simply no way an agency could predict
with confidence that a merger that would not significantly harm competition would nevertheless
facilitate some kind of future exclusionary conduct that would turn the merged firm into a monopolist. Indeed, it is far from clear that the Agencies can reliably predict even a merger’s short-term
price effects. The lack of empirical support for the accuracy of merger review or the efficacy of
merger enforcement is one of the great challenges presently facing antitrust scholars, economists,
and lawyers alike.11
An Unnecessary Task. Regardless of whether this kind of “potential future exclusionary conduct” analysis is possible, it is unnecessary. If a reviewing agency were to conclude with the req-
10
See, e.g., Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 767–68 (1984) (noting that “it is sometimes difficult to distinguish
robust competition from conduct with long-run anti-competitive effects”); Frank H. Easterbrook, When Is It Worthwhile to Use Courts to
Search for Exclusionary Conduct?, 2003 C OLUM . B US . L. R EV. 345, 345 (“The big problem lies in this: competitive and exclusionary conduct look alike.”); Einer Elhauge, Defining Better Monopolization Standards, 56 S TAN . L. R EV. 253, 255 (2003) (“[F]or decades monopolization doctrine has been governed by standards that are not just vague but vacuous.”); Kenneth L. Glazer & Abbott B. Lipsky, Jr., Unilateral
Refusals to Deal Under Section 2 of the Sherman Act, 63 A NTITRUST L.J. 749, 749 (1995) (“Courts have long struggled to define the specific forms of business conduct that constitute monopolization under Section 2 of the Sherman Act.”). Underscoring this uncertainty, by
our highly unscientific count, in the last few years the Antitrust Law Journal has devoted no less than three symposia issues (plus many
standalone articles) to the intense academic debate over how to identify exclusionary conduct. See Symposium—Aspen Skiing 20 Years
Later, 73 A NTITRUST L.J. 59 (2005); Symposium—Identifying Exclusionary Conduct Under Section 2, 73 A NTITRUST L.J. 311 (2005);
Symposium: Issues at the Forefront of Monopolization and Abuse of Dominance, 76 A NTITRUST L.J. 653 (2010).
11
See, e.g., Dennis W. Carlton, Why We Need to Measure the Effect of Merger Policy and How to Do It, CPI, Spring 2009, available at
https://www.competitionpolicyinternational.com/why-we-need-to-measure-the-effect-of-merger-policy-and-how-to-do-it/; Orley Ashenfelter,
Daniel Hosken & Matthew Weinberg, Generating Evidence to Guide Merger Enforcement, CPI, Spring 2009, available at https://www.
competitionpolicyinternational.com/generating-evidence-to-guide-merger-enforcement/.
the antitrust source
䡵 www.antitrustsource.com
䡵 October 2010
4
uisite certainty that the proposed transaction would give rise—on any coherent theory and in any
reasonable time-frame—to market power that outweighed any efficiencies that it generated, then
orthodox Section 7 enforcement would provide all the tools needed to block the merger or impose
the necessary remedies. And if the merged firm were to behave anticompetitively in the future,
there is no reason to think that the existing Section 2 and FTC Act Section 5 toolkit would not be
up to the task. While litigating exclusionary conduct is difficult, much of that difficulty lies in determining the merits of the claims—a difficulty that is exacerbated, not reduced, when the analysis
While litigating
is conducted ex ante rather than ex post. Similarly, while Section 2 remedies can be challenging
to develop and implement,12 that task is certainly not more difficult after-the-fact—when the con-
exclusionary conduct
duct has actually happened and been judged unlawful—than before-the-fact.13
is difficult, much of
worse than unnecessarily muddying the waters of an academic debate, then it might not be so
A Costly Task. If adding this “pre-emptive Section 2” analysis to merger review did nothing
objectionable. But we believe that expanding merger review to include prospective Section 2
that difficulty lies in
enforcement would come with a high price tag—actually, at least five high price tags.
determining the merits
every day. Adding an inquiry into hypothetical exclusionary conduct to merger review could sig-
First, merger review is already staggeringly expensive and intrusive, and growing more so
nificantly expand the scope and burden of that process.
Second, not only does “exclusionary” conduct often look like procompetitive conduct, it com-
of the claims—a
monly confers concrete benefits on consumers. Many types of “exclusionary” conduct under
difficulty that is
Section 2 involve the reduction of prices, at least in the short term; for example, it is only by careful analysis that one can target harmful predatory pricing without discouraging beneficial price
exacerbated, not
competition. Explicitly conditioning merger approval on an agency’s view of whether hypothetical
reduced, when the
petition could deter firms from considering (or implementing) such options even when consumers
future price cuts, bundled discounts, exclusive contracts, and so forth would threaten future comwould benefit from them.
Third, for similar reasons, blocking or imposing conditions on a merger because of such con-
analysis is conducted
cerns could actually harm competition—not merely by preventing the conduct itself but also by
ex ante rather than
sacrificing all or part of the benefits of the merger. Recall that the mergers affected by the change
would be those that would pass muster under “traditional” Section 7 analysis (i.e., mergers that
ex post.
are procompetitive or competitively neutral overall), but which nevertheless raise the specter of
hypothetical conduct that might at some future time lead to exclusion and to competitive harm.
Opposing such mergers because of the fear of future exclusion could kill the procompetitive bird
in the hand in the mere hope of catching an anticompetitive bird that might not be in the bush
at all.
Fourth, the new test increases the opportunities for competitive “gaming” of the review process.
Competitors already have strong incentives to oppose procompetitive mergers, and they now
have a whole new way to do so by advancing speculative theories of possible exclusionary conduct to the Agencies. This is not merely a hypothetical concern: the new Guidelines directly call
for competitor input on the risk of future exclusionary conduct.14
12
See, e.g., Symposium: Remedies for Dominant Firm Misconduct, 76 A NTITRUST L.J. 1 (2009); Renata B. Hesse, Section 2 Remedies and
U.S. v. Microsoft: What Is to Be Learned?, 75 A NTITRUST L.J. 847 (2009); David A. Heiner, Single-Firm Conduct Remedies: Perspectives
from the Defense, 75 A NTITRUST L.J. 871 (2009).
13
Only in science fiction do we punish “offenders” for offenses that they have not yet committed. Compare M INORITY R EPORT (DreamWorks/20th Century Fox, 2002).
14
2010 Guidelines, supra note 1, § 2.2.3.
the antitrust source
䡵 www.antitrustsource.com
䡵 October 2010
5
Fifth, the speculative nature of exclusion analysis would exacerbate the general reduction in
clarity in the new Guidelines. The 2010 revisions move away from clear declarations of agency
practice and towards a more flexible and indeterminate description of factors that vary in significance from one case to the next. While the new text may indeed more accurately reflect agency
practice, the change in style could undermine the usefulness of the Guidelines by making them
less easily applicable by courts and less helpful to merging parties, and by reducing the “gravitational” unifying effect on agency practice that a clear, simple roadmap can offer.15
Conclusion
There is no reason to think that existing law cannot adequately address concerns about exclusionary conduct that may arise from a proposed merger. Nor is there any reason to think that
adding a speculative, hypothetical version of Section 2 analysis to merger enforcement will
improve the benefits or lessen the burdens of merger review in any way. There is much reason to
fear that Section 2’s uncertainties, injected into merger review, will make that review more costly,
more vulnerable to gamesmanship, less predictable, and less effective. The Agencies should
tread this new and shaky ground with great caution. 䢇
15
See, e.g., D. Bruce Hoffman, How Much Discretion Is Enough?, T HE D EAL (May 27, 2010), available at http://www.thedeal.com/
newsweekly/community/judgment-call/how-much-discretion-is-enough.php.